Na+/K+-ATPase with a Blocked E1ATP Site Still Allows Backdoor Phosphorylation of the E2ATP site

Abstract

The role of simultaneously existing ATP‐binding sites in the catalytic process of Na+/K+‐ATPase is unclear. In order to learn whether blocking the E1 ATP site affects the properties of the E2 ATP site, the E1 ATP site was inactivated by either fluorescein 5′‐isothiocyanate, the non‐phosphorylating Cr(H2O)4‐AdoPP [CH2]P or the phosphorylating Cr(H2O)4ATP. The properties of the remaining E2ATP site were studied by measuring ‘backdoor phosphorylation’ in the presence of ouabain, or K+‐activated hydrolysis of p‐nitrophenyl phosphate. The involvement of the E2ATP site was further tested by the effects of Co(NH3)4ATP, a specific inactivator of this site. When the E1ATP site was inactivated by fluorescein 5′‐isothiocyanate or the non‐phosphorylating Cr(H2O)4AdoPP [CH2]P, backdoor phosphorylation and the activity of K+‐activated p‐nitrophenylphosphatase remained unchanged. Both processes were lost, however, when the E2ATP site was additionally inactivated by Co(NH3)4ATP. Inactivation of the E1ATP site by fluorescein 5′‐isothiocyanate or Cr(H2O)4 AdoPP [CH2]P decreased the affinity of the P‐nitrophenyl‐phosphatase activity of the E2ATP site for the substrate p‐nitrophenyl phosphate by four times. This is consistent with a former report showing that dephosphorylation in a fluorescein 5′‐isothiocyanate‐inactivated Na+/K+‐ATPase has a lowered sensitivity for ATP [Scheiner‐Bobis, G., Antonipillai, J. & Farley, R. A. (1993) Biochemistry 32, 9592–9599]. Inactivation of the E1ATP site by the phosphorylating Cr(H2O)4ATP, however, led to a loss of the property of the E2ATP site to hydrolyse K+‐dependent p‐nitrophenyl phosphate and to achieve backdoor phosphorylation. Evidently, ATP sites coexist in Na+/K+‐ATPase, and binding of ATP to one site affects the property of the other site [Scheiner‐Bobis, G., Esmann, M. & Schoner, W. (1989) Eur. J. Biochem. 183, 173–178]. Although the enzyme can be phosphorylated from both ATP sites, phosphorylation of the E1ATP site excludes the phosphorylation of the E2ATP site.

A Two‐Site Model of Interacting ATP Sitesa

“…So far, these are the conversion of the E 1 ATP site into a Na + ‐form when the E 2 ATP site is blocked by Co(NH 3 ) 4 ATP 13 and the inhibition of backdoor phosphorylation (of the E 2 ATP binding site) by phosphorylation of the E 1 ATP binding site(FIG. 6) 6,7 Furthermore, we should keep in mind that “ES” in the two‐site model (F ig . 4) does not discriminate between the possibilities that ATP is merely bound or that it exists as ADP + phosphointermediate.…”

The phosphatase activity of the isolated H4‐H5 loop of Na+/K+ ATPase resides outside its ATP binding site

“…The observation of high and low affinity ATP sites with approximate K d values of 1 µ m (E 1 ATP site) and 200 µ m (E 2 ATP site) in the membrane‐embedded Na + /K + ATPase, and the finding that reaction inert MgATP complex analogues such as Cr(H 2 O) 4 ATP and Co(NH 3 ) 4 ATP may react specifically with these ATP sites [6] as well as the complex kinetics with the fluorescent 2′(3′)‐ O ‐(6‐ N ′, N ′‐dimethylaminonaphthalenesulfonyl)ATP (DANSyl‐ATP) [7,8], lead to the suggestion that high and low affinity ATP sites coexist and that they interact during catalysis. Consistent with this conclusion is the finding that the activity of a K + activated phosphatase, which represents a partial function of the ATP site [9–12], was almost unaffected by the blockade of the ATP site due to modification of K501 with fluorescein isothiocyanate (FITC) [13], but was lost when the enzyme additionally reacted with Co(NH 3 ) 4 ATP [14,15] or erythrosin isothiocyanate [16]. The latter binds to C549 within the nucleotide (N)‐domain of Na + /K + ATPase [16].…”

Affinity labelling with MgATP analogues reveals coexisting Na+ and K+ forms of the α‐subunits of Na+/K+‐ATPase

“…A number of studies seem to refute the formerly favoured idea that Na + /K + pumping is the result of the conformational change of a single catalytic α‐subunit and its single ATP binding site as proposed by the so‐called Albers–Post model. Hence, this model excludes simultaneously existing and cooperating ATP binding sites [2,3] and the finding that specific labelling of the E 1 ATP or the E 2 ATP binding sites does not block labelling and partial activities of the other remaining empty site [4–9]. In addition, the demonstration of ‘superphosphorylation’, i.e.…”

“…Tryptic digestion of Cr(H 2 O) 4 AdoPP[CH 2 ]P‐inactivated Na + /K + ‐ATPase. Cr(H 2 O) 4 Ado PP [CH 2 ] P inactivates Na + /K + ‐ATPase by forming a stable E 1 ATP complex [5,6]. The MgATP‐complex analogue is unable to phosphorylate the enzyme.…”

“…The MgATP‐complex analogue is unable to phosphorylate the enzyme. Binding to the E 1 ATP binding site does not affect the partial activities of the E 2 conformation such as occlusion of 86 Rb + , [ 3 H]ouabain binding by backdoor phosphorylation and K + ‐activated p ‐nitrophenylphosphatase [5,6]. When Cr(H 2 O) 4 ‐Ado PP [CH 2 ] P ‐inactivated Na + /K + ‐ATPase was treated with trypsin, C‐terminal 80‐kDa fragments resembling the Na + conformation, C‐terminal 60‐kDa and N‐terminal 40‐kDa peptides resembling the K + conformation [22,23] were simultaneously seen (Fig.…”